Abstract
We have previously established that saccadic reaction times (SRTs) may depend on reinforcement contingencies. It follows that one must be able to discriminate one's own latencies to adequately assign credit to one's actions. In two different experiments we used an adaptive procedure to test the limit of both saccadic and manual reaction time (MRTs) in ten subjects for each experiments. In the first experiment we trained ten participants using a staircase paradigm to determine their 75% perceptual threshold of SRTs. On each trial they had to saccade to a stepping target. In a 2-AFC task they had to choose the number representing the actual SRT while the second number was a made-up value which proportionally differed from this SRT. The relative difference between the two options was computed by either adding or subtracting one of the percentage values of a decreasing fixed staircase range. To encourage learning a feedback was provided after each response. In the MRTs experiment participants reached a target and judged their reaction time in a 2AFC task. The protocol was similar to the SRTs experiment, except that we used a double staircase in a QUEST procedure. In this case the relative difference between the two numbers quickly converge to threshold. 75% threshold was computed by fitting a psychometric function for both experiments. Results reveal a very accurate perception both of SRTs and MRTs: 75% threshold range from 13% for the best performance to the 37% for the worst one in the SRTs discrimination and from 12% to 36% for MRTs perception. This indicates that our participants can discriminate very small SRT differences, providing support for the possibility that the credit assignment problem may be solved even for short reaction times.
Meeting abstract presented at VSS 2018